Anonymization Standards

Asked Jul 01 '21 at 13:38

Active Jul 02 '21 at 13:15

Viewed 85 times

I'm interested in data anonymization techniques and scores for the quality of data anonymization in "the real world". I was told this is also part of cryptography.

What I know: In theory one can use k-anonymity, l-diversity or t-closeness to define the quality. Also the open gda score could be used. Common techniques are generalization, synthetic data and deleting data. I also studied the laws and found out, that there are no strict rules which state a standard for anonymization (I only found out, that the EuGH once said that reidentification should be "hard")

My question: How is anonymization done in practice? What standards are common? Which scores for quality are used and why?

edited Jul 02 '21 at 13:15

asked Jul 01 '21 at 13:38

Titanlord

2,244
11
31

1

Anonymity doesn't make sense without any context. I think you should precise the use-case. – Ievgeni Jul 01 '21 at 14:58
The main context I found in the literature is data in tabular form. For example csv or xlxx files. I don't mean anonymization done by VPN, Proxies, the Onion network or anything like that. I think the correct term for what I'm interested in is "data anonymization" (I added that to the question) – Titanlord Jul 02 '21 at 13:14
And/or differential privacy? – Paul Uszak Jul 02 '21 at 16:41
@PaulUszak Is it a standard? – Ievgeni Jul 04 '21 at 16:29

Anonymization Standards

0 Answers0